| 摘要 |
PURPOSE:To prevent the generation of granular noise in a high-light section and thereby make possible the visualization of a high-quality color image with superior gradation and resolution by quantizing each color image signal using an error diffusion method when a detected achromatic component exceeds a specified value. CONSTITUTION:Quantization is performed by an adder 6 and a comparative subtractor 8 using an error diffusion method. An output from a selection circuit 4 is entered into not only a comparator 5 but also the adder 6. In the adder 6, an error between input data which generates with previous pixels which are output from the comparative subtractor 8 and output data, and the density data of outstanding pixels are added together through the timing of clock pulse. After this, the data added together is output to the comparative subtractor 8. The comparative substractor 8 compares the output from the adder 6 as an input with a threshold value '80H' through the timing of clock pulse. If the input data value is larger than 80H, '1' is output to the selection circuit 11 by the comparative substractor 8, while if the input data value is smaller than or equal to 80H, '0' is output. |
